The quick impact of Generative AI and LLMs has caught SaaS off guard.
- Most SaaS companies don't have substantial technical competitive advantage
- Switching costs are in fact, relatively low in specific categories ( you can export your sales data, for example)
- SaaS does not inherently take advantage of legacy on-premise data assets, or at least not without some messy integration which is contrary to the SaaS Business model. Yes, there are "connectors" and EAI (Enterprise Application Integration) workflows; these have existed as far back as Informatica and Tibco, but many SaaS companies have neglected building/integrating these capabilities in the hope that wholesale migration to their platforms would take place instead. A clear exception here is Microsoft Azure, which was designed with integration between on-premise and cloud assets in mind.
- The "harvesting of future cash flows" aspect of SaaS runs counter to the "discovery of insights based on correlations found" with AI and on-premise systems. Much of SaaS was developed around a thesis of labor arbitrage, not computing breakthroughs. That worked for two decades in a relatively linear world, but AI now inserts a step function that SaaS companies will struggle to keep up with.
Prediction: There will be a massive switch of market cap, with SaaS companies deflating while AI companies inflating. This is not dissimilar to the switch from client/server enterprise software to SaaS and internet-accessible applications and, before that, from mainframe computing to client/server.
Comments
You can follow this conversation by subscribing to the comment feed for this post.